You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Joel Bernstein (JIRA)" <ji...@apache.org> on 2015/06/30 17:04:04 UTC
[jira] [Comment Edited] (SOLR-7441) Improve overall robustness of
the Streaming stack: Streaming API, Streaming Expressions, Parallel SQL
[ https://issues.apache.org/jira/browse/SOLR-7441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14608428#comment-14608428 ]
Joel Bernstein edited comment on SOLR-7441 at 6/30/15 3:03 PM:
---------------------------------------------------------------
Added a patch that adds an Exception handling framework.
The new ExceptionStream wraps a TupleStream and catches any exceptions from the underlying Stream. It then logs the exception and returns a Tuple with the exception message included.
The SolrStream has some code added to look for Tuples with an
Exception message and then to throw an Exception passing along the exception message from the Tuple.
This effectively allows exceptions to be passed through any number of distributed tiers.
The nice thing about this design is that the SolrStream throws and the ExceptionStream catches. All other Streams in between can ignore exception handling entirely.
No tests yet to prove this concept actually works, but it looks promising.
was (Author: joel.bernstein):
Added a patch that adds an Exception handling framework.
The new ExceptionStream wraps a TupleStream and catches any exceptions from the underlying Stream. It then logs the exception and returns a Tuple with the exception message included.
The SolrStream has some code added to look for Tuples with an
Exception message and then to throw an Exception passing along the exception message from the Tuple.
This effectively allow exceptions to be passed through any number of distributed tiers.
The nice thing about this design is that the SolrStream throws and the ExceptionStream catches. All other Streams in between can ignore exception handling entirely.
No tests yet to prove this concept actually works, but it looks promising.
> Improve overall robustness of the Streaming stack: Streaming API, Streaming Expressions, Parallel SQL
> -----------------------------------------------------------------------------------------------------
>
> Key: SOLR-7441
> URL: https://issues.apache.org/jira/browse/SOLR-7441
> Project: Solr
> Issue Type: Improvement
> Affects Versions: 5.1
> Reporter: Erick Erickson
> Assignee: Joel Bernstein
> Priority: Minor
> Attachments: SOLR-7441.patch, SOLR-7441.patch, SOLR-7441.patch
>
>
> It's harder than it could be to figure out what the error is when using Streaming Aggregation. For instance if you specify an fl parameter for a field that doesn't exist it's hard to figure out that's the cause. This is true even if you look in the Solr logs.
> I'm not quite sure whether it'd be possible to report this at the client level or not, but it seems at least we could repor something more helpful in the Solr logs.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org